RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation
Despite the progress on 3D point cloud deep learning, most prior works focus on learning features that are invariant to translation and point permutation, and very limited efforts have been devoted for rotation invariant property. Several recent studies achieve rotation invariance at the cost of low...
Saved in:
Main Authors: | , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2024
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/9747 https://ink.library.smu.edu.sg/context/sis_research/article/10747/viewcontent/2408.06110v1.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-10747 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-107472024-12-16T03:15:47Z RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation ZHANG, Zhiyuan YANG, Licheng XIANG Zhiyu, Despite the progress on 3D point cloud deep learning, most prior works focus on learning features that are invariant to translation and point permutation, and very limited efforts have been devoted for rotation invariant property. Several recent studies achieve rotation invariance at the cost of lower accuracies. In this work, we close this gap by proposing a novel yet effective rotation invariant architecture for 3D point cloud classification and segmentation. Instead of traditional pointwise operations, we construct local triangle surfaces to capture more detailed surface structure, based on which we can extract highly expressive rotation invariant surface properties which are then integrated into an attention-augmented convolution operator named RISurConv to generate refined attention features via self-attention layers. Based on RISurConv we build an effective neural network for 3D point cloud analysis that is invariant to arbitrary rotations while maintaining high accuracy. We verify the performance on various benchmarks with supreme results obtained surpassing the previous state-of-the-art by a large margin. We achieve an overall accuracy of 96.0{\%} (+4.7{\%}) on ModelNet40, 93.1{\%} (+12.8{\%}) on ScanObjectNN, and class accuracies of 91.5{\%} (+3.6{\%}), 82.7{\%} (+5.1{\%}), and 78.5{\%} (+9.2{\%}) on the three categories of the FG3D dataset for the fine-grained classification task. Additionally, we achieve 81.5{\%} (+1.0{\%}) mIoU on ShapeNet for the segmentation task. 2024-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/9747 info:doi/10.1007/978-3-031-73390-1_6 https://ink.library.smu.edu.sg/context/sis_research/article/10747/viewcontent/2408.06110v1.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Point cloud Rotation invariant Attention Deep learning Artificial Intelligence and Robotics Computer Sciences |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Point cloud Rotation invariant Attention Deep learning Artificial Intelligence and Robotics Computer Sciences |
spellingShingle |
Point cloud Rotation invariant Attention Deep learning Artificial Intelligence and Robotics Computer Sciences ZHANG, Zhiyuan YANG, Licheng XIANG Zhiyu, RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
description |
Despite the progress on 3D point cloud deep learning, most prior works focus on learning features that are invariant to translation and point permutation, and very limited efforts have been devoted for rotation invariant property. Several recent studies achieve rotation invariance at the cost of lower accuracies. In this work, we close this gap by proposing a novel yet effective rotation invariant architecture for 3D point cloud classification and segmentation. Instead of traditional pointwise operations, we construct local triangle surfaces to capture more detailed surface structure, based on which we can extract highly expressive rotation invariant surface properties which are then integrated into an attention-augmented convolution operator named RISurConv to generate refined attention features via self-attention layers. Based on RISurConv we build an effective neural network for 3D point cloud analysis that is invariant to arbitrary rotations while maintaining high accuracy. We verify the performance on various benchmarks with supreme results obtained surpassing the previous state-of-the-art by a large margin. We achieve an overall accuracy of 96.0{\%} (+4.7{\%}) on ModelNet40, 93.1{\%} (+12.8{\%}) on ScanObjectNN, and class accuracies of 91.5{\%} (+3.6{\%}), 82.7{\%} (+5.1{\%}), and 78.5{\%} (+9.2{\%}) on the three categories of the FG3D dataset for the fine-grained classification task. Additionally, we achieve 81.5{\%} (+1.0{\%}) mIoU on ShapeNet for the segmentation task. |
format |
text |
author |
ZHANG, Zhiyuan YANG, Licheng XIANG Zhiyu, |
author_facet |
ZHANG, Zhiyuan YANG, Licheng XIANG Zhiyu, |
author_sort |
ZHANG, Zhiyuan |
title |
RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
title_short |
RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
title_full |
RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
title_fullStr |
RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
title_full_unstemmed |
RISurConv : Rotation invariant surface attention-augmented convolutions for 3D point cloud classification and segmentation |
title_sort |
risurconv : rotation invariant surface attention-augmented convolutions for 3d point cloud classification and segmentation |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2024 |
url |
https://ink.library.smu.edu.sg/sis_research/9747 https://ink.library.smu.edu.sg/context/sis_research/article/10747/viewcontent/2408.06110v1.pdf |
_version_ |
1819113126486867968 |